CN109670676A - Distributing net platform region method for prewarning risk and system based on Support Vector data description - Google Patents
Distributing net platform region method for prewarning risk and system based on Support Vector data description Download PDFInfo
- Publication number
- CN109670676A CN109670676A CN201811418180.XA CN201811418180A CN109670676A CN 109670676 A CN109670676 A CN 109670676A CN 201811418180 A CN201811418180 A CN 201811418180A CN 109670676 A CN109670676 A CN 109670676A
- Authority
- CN
- China
- Prior art keywords
- platform region
- net platform
- training sample
- support vector
- vector data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000013598 vector Substances 0.000 title claims abstract description 84
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000012549 training Methods 0.000 claims abstract description 88
- 238000009826 distribution Methods 0.000 claims abstract description 59
- 238000000513 principal component analysis Methods 0.000 claims abstract description 15
- 230000007613 environmental effect Effects 0.000 claims description 13
- 238000010276 construction Methods 0.000 claims description 10
- 230000009467 reduction Effects 0.000 claims description 10
- 238000004458 analytical method Methods 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 5
- RVRCFVVLDHTFFA-UHFFFAOYSA-N heptasodium;tungsten;nonatriacontahydrate Chemical compound O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.O.[Na+].[Na+].[Na+].[Na+].[Na+].[Na+].[Na+].[W].[W].[W].[W].[W].[W].[W].[W].[W].[W].[W] RVRCFVVLDHTFFA-UHFFFAOYSA-N 0.000 claims description 5
- 238000013139 quantization Methods 0.000 claims description 5
- 230000009466 transformation Effects 0.000 claims description 5
- 238000009434 installation Methods 0.000 claims description 4
- 238000005303 weighing Methods 0.000 claims description 3
- 238000012706 support-vector machine Methods 0.000 claims 1
- 230000000694 effects Effects 0.000 abstract description 4
- 238000005516 engineering process Methods 0.000 abstract 1
- 238000012502 risk assessment Methods 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 5
- 238000002474 experimental method Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 238000011156 evaluation Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000004140 cleaning Methods 0.000 description 2
- 235000013399 edible fruits Nutrition 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000001627 detrimental effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000012847 principal component analysis method Methods 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 230000032258 transport Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Evolutionary Computation (AREA)
- General Business, Economics & Management (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Tourism & Hospitality (AREA)
- Life Sciences & Earth Sciences (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of distributing net platform region method for prewarning risk and system based on Support Vector data description belongs to big data and excavates and applied technical field, including obtaining the operation risk factor of distributing net platform region as training sample set；Dimension-reduction treatment is carried out to the training sample in training sample set using Principal Component Analysis, obtains the training sample of low dimensional；The feature difference of training sample based on low dimensional, the building weighting more disaggregated models of Support Vector data description carry out Classification and Identification to the operating status of power distribution station, to realize the Risk-warning of the distributing net platform region.The present invention is by constructing power distribution station operation risk Early-warning Model using big data technology, and risk assessment is high-efficient, early warning effect is good.
Description
Technical field
It is excavated the present invention relates to big data and applied technical field, in particular to it is a kind of based on Support Vector data description
Distributing net platform region method for prewarning risk and system.
Background technique
With the construction and development of China's smart grid, power supply enterprise has accumulated the power distribution network operation data of magnanimity, but passes
The reliability estimation method of system is limited by data source, Data Integration, data-handling capacity, is difficult multi-source heterogeneous from distribution system
The information being more worth is excavated in data.
Conventional evaluating reliability of distribution network be mainly based upon distribution system for a long time under different operating statuses reliability system
Evaluation, based on cutting load index, appraisal procedure is mainly analytic method and simulation, is as a result mainly used for distribution network planning and sets
Meter, though it can be considered that probability and consequence that failure occurs, its measurement period is long, and time scale is big, generally several months or number
Year, it is difficult to directly apply to the assessment of distributing net platform region short term reliability and real time execution scheduling controlling.
In addition, the replacement of distributing net platform region transformer equipment or retired expection are mainly according to Variant number, the operation time limit, history
A small amount of factor such as peak load or overall life cycle cost determine, lack objective evaluation.Power distribution network transformer equipment state
Judgement depends on prevention experiment and periodic inspection to realize, and the more specific parameter concentrated on to electric appliance or mechanical aspects
It carries out checking the state analysis based on single or a small amount of parameter with some macroscopic views.
Summary of the invention
The distributing net platform region method for prewarning risk that the purpose of the present invention is to provide a kind of based on Support Vector data description and
System, to improve the accuracy of distributing net platform region Risk-warning.
In order to achieve the above object, the present invention uses a kind of distributing net platform region Risk-warning side based on Support Vector data description
Method includes the following steps:
The operation risk factor of distributing net platform region is obtained as training sample set, which includes and transformer
Relevant risks and assumptions, to the risks and assumptions of operation of power networks environmental correclation and risks and assumptions relevant with external environment；
Dimension-reduction treatment is carried out to the training sample in the training sample set using Principal Component Analysis, obtains low dimensional
Training sample；
The feature difference of training sample based on low dimensional, the building weighting more disaggregated models of Support Vector data description；
According to the weighting more disaggregated models of Support Vector data description, feature is carried out to the operation risk factor of distributing net platform region and is mentioned
It takes, identifies the power distribution station of different conditions, to realize the Risk-warning of the distributing net platform region.
Preferably, the risks and assumptions relevant to transformer include transformer state amount factor R1, transformer station high-voltage side bus year
Limit factor R2And transformer installation site height above sea level impact factor R3, in which:
R2=eN-M,
R3=e(h-1000)/8150,
In formula, WiIndicate that equipment i relevant with transformer scores recently, aiIndicate equipment i relevant with transformer to transformation
The weighing factor of device, n indicate equipment sum relevant with transformer；N indication transformer actually puts into operation the time, and M is and transformation
The specified maximum runing time of device, e is Euler's numbers.
Preferably, described and operation of power networks environmental correclation risks and assumptions include power distribution station load factor factor L and distribution
Platform area three-phase imbalance factor T, power distribution station load factor factor L include heavy-overload time L in the transformer unit time1, unit
Heavy-overload extreme value L in time2And heavy-overload severity L in the unit time3；
Distributing net platform region three-phase imbalance factor T includes the three-phase imbalance time T of transformer1, the three-phase of transformer it is uneven
Weigh extreme value T2With the three-phase imbalance severity T of transformer3。
Preferably, the risks and assumptions relevant to external environment include ambient weather state factor W, ambient weather state
Factor W includes the temperature, wind-force and weather condition of environment described in power distribution station.
Preferably, described that dimension-reduction treatment is carried out to the training sample using Principal Component Analysis, obtain the instruction of low dimensional
Practice sample, comprising:
By the risks and assumptions and and external rings of the risks and assumptions relevant to transformer and operation of power networks environmental correclation
The relevant risks and assumptions in border form multidimensional characteristic vectors；
Dimension-reduction treatment is carried out to the multidimensional characteristic vectors using Principal Component Analysis, obtains the training sample of low dimensional.
Preferably, the feature difference of the training sample based on low dimensional, building weighting Support Vector data description
Model, comprising:
Percentage contribution of the training sample based on the low dimensional when constructing Support Vector data description model, to low-dimensional
Corresponding weighted value is arranged in the training sample of degree；
By in the training sample of the low dimensional and its corresponding weighted value input Support Vector data description model, construct
The weighting Support Vector data description model.
It preferably, include the sample set of multiple classifications, the training based on low dimensional in the training sample set
The feature difference of sample, building weighting Support Vector data description model, comprising:
With the sample set of each classification in the training sample set, the sample set for constructing each classification respectively is corresponding
It is minimum surround suprasphere, as the corresponding weighting Support Vector data description model of each classification, the weighting supporting vector
Data descriptive model are as follows:
Wherein, | | xi-c||2≤R2+ξi, ξi>=0, R indicate that suprasphere radius, C indicate that penalty coefficient, c indicate suprasphere ball
The heart, siIndicate weight coefficient, ξiIndicate slack variable, xiIndicate that training sample, n indicate training samples number.
Preferably, described that spy is carried out according to the operation risk factor of the weighting Support Vector data description model to distributing net platform region
Sign is extracted, and realizes the Risk-warning of the distributing net platform region, comprising:
Using the corresponding weighting Support Vector data description model of each classification, the operation of the distributing net platform region is calculated
Relative distance of the risks and assumptions to the corresponding suprasphere of each classification；
Compare the operation risk factor of the distributing net platform region to the corresponding suprasphere of each classification relative distance size,
And export classification corresponding to the suprasphere with lowest distance value as a result, realize the Risk-warning of distributing net platform region.
Preferably, after the operation risk factor for obtaining distributing net platform region is as training sample set, further includes:
The operation risk factor of the distributing net platform region is pre-processed, at the operation risk factor of the distributing net platform region
Manage into the quantization index value that can be used for modeling analysis and mode input.
On the other hand, a kind of distributing net platform region Warning System based on Support Vector data description is provided, comprising: obtain
Module, dimensionality reduction module, model construction module and Risk-warning module；
It obtains module and is used to obtain the operation risk factor of distributing net platform region as training sample set, the operation risk factor
Including risks and assumptions relevant to transformer, with the risks and assumptions of operation of power networks environmental correclation and wind relevant with external environment
The dangerous factor；
Dimensionality reduction module is used to carry out at dimensionality reduction the training sample in the training sample set using Principal Component Analysis
Reason, obtains the training sample of low dimensional；
Model construction module is used for the feature difference of the training sample based on low dimensional, building weighting supporting vector data
More disaggregated models are described；
Risk-warning module is used for according to the weighting more disaggregated models of Support Vector data description, to the operation wind of distributing net platform region
The dangerous factor carries out feature extraction, the power distribution station of different conditions is identified, to realize the Risk-warning of the distributing net platform region.
Compared with prior art, there are following technical effects by the present invention: the present invention is in order to solve distributing net platform region running environment
The problem of complicated, operation risk is difficult to rapid evaluation, from transformer itself, operation of power networks environment and distributing net platform region external environment three
A aspect has evaluated distributing net platform region and runs relevant risks and assumptions, carries out dimensionality reduction to risks and assumptions using principal component analysis method,
It is then based on the feature difference of risks and assumptions after dimensionality reduction, building weighting Support Vector data description model transports distributing net platform region
Row risk carries out early warning.The present invention by power distribution network running environment is complicated, load is changeable, the unusual fluctuation time is multiple the features such as, answer
With electric power big data scientific discovery power distribution network moving law with the generation of all kinds of unusual fluctuation events of accurate early warning, examined for distribution O&M
It repairs, the improvement of Study on Power Grid Planning, power distribution network reliably assists supporting.
Detailed description of the invention
With reference to the accompanying drawing, specific embodiments of the present invention will be described in detail:
Fig. 1 is a kind of flow diagram of distributing net platform region method for prewarning risk based on Support Vector data description；
Fig. 2 is distributing net platform region data classification schematic diagram；
Fig. 3 is power distribution station risk identification procedure chart；
Fig. 4 is the flow diagram integrated to risk related data；
Fig. 5 is the classification knot that the relevant risks and assumptions of substation equipment and the distributing net platform region load factor factor are obtained as input
Fruit schematic diagram；
Fig. 6 is that the relevant risks and assumptions of substation equipment are obtained with the distribution net platform region three-phase imbalance factor as input
Classification results schematic diagram；
Fig. 7 is the classification knot that the distribution net platform region three-phase imbalance factor and the distributing net platform region load factor factor are obtained as input
Fruit schematic diagram；
Fig. 8 is a kind of structural schematic diagram of distributing net platform region Warning System based on Support Vector data description.
Specific embodiment
In order to further explain feature of the invention, reference should be made to the following detailed description and accompanying drawings of the present invention.Institute
Attached drawing is only for reference and purposes of discussion, is not used to limit protection scope of the present invention.
As shown in Figure 1, present embodiment discloses a kind of distributing net platform region Risk-warning side based on Support Vector data description
Method includes the following steps S1 to S4:
S1, obtain distributing net platform region the operation risk factor as training sample set, which includes and change
The relevant risks and assumptions of depressor, to the risks and assumptions of operation of power networks environmental correclation and risks and assumptions relevant with external environment；
It should be noted that according to the difference with distribution net platform region degree of correlation, by the related data of distribution net platform region point
For two class of core data and noncore data, core data refers to the data being directly linked with the operation of distribution net platform region, that is, passes through
What equipment itself or power grid sensor directly acquired, and the data of energy directly consersion unit state.Noncore data is and equipment
The data of indirect association influence the data of equipment by other media, and cannot direct consersion unit state.As shown in Fig. 2, core
Calculation is according to including equipment class data, operation class data and measurement class data.Noncore data includes GEOGRAPHIC ATTRIBUTES data and Meteorological Change
Data.
S2, dimension-reduction treatment is carried out to the training sample in the training sample set using Principal Component Analysis, obtained low
The training sample of dimension；
The feature difference of S3, training sample based on low dimensional, building weighting Support Vector data description are classified mould more
Type；
S4, according to the weighting more disaggregated models of Support Vector data description, the operation risk factor of distributing net platform region is carried out special
Sign is extracted, and the power distribution station of different conditions is identified, to realize the Risk-warning of the distributing net platform region.
It should be noted that considering training sample in construction Support Vector data description in this programme when constructing model
The difference of percentage contribution when (support vector data description, SVDD) minimum sphere body Model introduces power
Weight coefficient is weighted training sample, and avoid causes training to obtain because training sample has differences to model contribution degree
Model inaccuracy problem, to improve model to the accuracy of distributing net platform region Risk-warning.
Preferably, the risks and assumptions relevant to transformer include transformer state amount factor R1, transformer station high-voltage side bus year
Limit factor R2And transformer installation site height above sea level impact factor R3, in which:
R2=eN-M,
R3=e(h-1000)/8150,
In formula, WiIndicate that equipment i relevant with transformer scores recently, aiIndicate equipment i relevant with transformer to transformation
The weighing factor of device, n indicate equipment sum relevant with transformer；N indication transformer actually puts into operation the time, and M is and transformation
The specified maximum runing time of device, the generalized time of transformer are typically no less than 30 years, and taking M is that 30, h is expressed as where transformer
Height above sea level, e is Euler's numbers, i.e. the truth of a matter of natural logrithm function, e=2.718281828459.
Preferably, described and operation of power networks environmental correclation risks and assumptions include power distribution station load factor factor L and distribution
Platform area three-phase imbalance factor T, power distribution station load factor factor L include heavy-overload time L in the transformer unit time1, unit
Heavy-overload extreme value L in time2And heavy-overload severity L in the unit time3；Distributing net platform region three-phase imbalance factor T includes becoming
The three-phase imbalance time T of depressor1, transformer three-phase imbalance extreme value T2With the three-phase imbalance severity T of transformer3。
Wherein:
L2=max (lv), 1≤v≤m,
In formula: m indicates the frequency of heavy-overload, tvFor the duration of the v times heavy-overload, lvIndicate the v times heavy-overload
In the overload degree of t moment, l indicates the averaged overload degree of all numbers.
T2=max (l 'u), 1≤u≤m`,
In formula, m` indicates the frequency of three-phase imbalance, tu` is three-phase imbalance duration, lu` is indicated in t`
The three-phase imbalance degree at quarter, TuIndicate the three-phase imbalance degree at the t` moment, △ T is power distribution station three-phase imbalance standard
Numerical value.
Preferably, the risks and assumptions relevant to external environment include ambient weather state factor W, ambient weather state
Factor W includes temperature (such as highest temperature, the lowest temperature), wind-force (such as maximum wind power) and the day of environment described in power distribution station
Gas situation.
Preferably, above-mentioned steps S2: dimension-reduction treatment is carried out to the training sample using Principal Component Analysis, obtains low-dimensional
The training sample of degree.It includes following subdivided step:
By the risks and assumptions and and external rings of the risks and assumptions relevant to transformer and operation of power networks environmental correclation
The relevant risks and assumptions in border form multidimensional characteristic vectors；
Dimension-reduction treatment is carried out to the multidimensional characteristic vectors using Principal Component Analysis, obtains the training sample of low dimensional.
The present embodiment reduces the information redundancy between each factor variable by dimension-reduction treatment, provides sorting algorithm below
Operational efficiency and accuracy.
Specifically, the relevant risk factor of distributing net platform region operation is extracted in the present embodiment, constructs 10 dimensional feature vector (R1,
R2,R3,L1,L2,L3,T1,T2,T3, W), by principal component analysis, dimension-reduction treatment is carried out to input vector, as distributing net platform region wind
Dangerous state feature description.Wherein, weight is removed from the relevant risk factor that distribution net platform region is run using principal component analytical method
Multiple or strong correlation influence factor, screening obtains influencing the key factor of platform area operation risk, to eliminate detrimental effect
Factor improves modeling efficiency.
Preferably, above-mentioned steps S4: the feature difference of the training sample based on low dimensional, building weighting supporting vector number
According to descriptive model.It includes following subdivided step:
Percentage contribution of the training sample based on the low dimensional when constructing Support Vector data description model, to low-dimensional
Corresponding weighted value is arranged in the training sample of degree；
By in the training sample of the low dimensional and its corresponding weighted value input Support Vector data description model, construct
The weighting Support Vector data description model.
The present embodiment is directed to different training sample xi, in construction Support Vector data description (SVDD) minimum sphere body Model
When percentage contribution be different, weighting Support Vector data description (WSVDD) algorithm introduce variable siTo different training samples
This work suitably weights, and avoids leading to the problem of training result inaccuracy because sample has differences model contribution degree.
It should be noted that weight coefficient siQuantization determination process are as follows:
In view of during SVDD Construction of A Model, the training sample in different densities region influences minimal hyper-sphere radius
Degree be obviously different, i.e. the sample of the high-density region sample that is better than density regions to the percentage contribution of model,
The weight coefficient method for solving based on training sample centre distance is used in the present embodiment:
Wherein: d (xi) it is Euclidean distance of the training sample point to center of a sample, davr, dmin, dmaxRespectively all training
The average value of center of a sample's distance, minimum value, maximum value.In addition ε is required sufficiently small, meets 0 < ε < 1.P >=2 and be positive integer.
It, can be according to hands-on situation adjustment parameter ε and p, to make to weight Support Vector data description when determining weight coefficient
(weighted support vector data description, WSVDD) i.e. WSVDD model is optimal.
Preferably, training sample the set { (xi, yi), i=1,2 ... ..., n } in include multiple classifications sample
Collect, wherein yj∈ { 1,2 ... ... N } is its class label, and n is number of training, and N is classification number.Above-mentioned steps S3: it is based on
The feature difference of the training sample of low dimensional, building weighting Support Vector data description model, includes the following steps:
With the sample set of each classification in the training sample set, structure is distinguished to N class sample set affiliated in sample
Corresponding WSVDD model is built, as the corresponding weighting Support Vector data description model of each classification, that is, is established N number of based on phase
The minimum of sample set is answered to surround suprasphere.The weighting Support Vector data description model are as follows:
Wherein, | | xi-c||2≤R2+ξi, ξi>=0, R indicate that suprasphere radius, C indicate that penalty coefficient, c indicate suprasphere ball
The heart, siIndicate weight coefficient, ξiIndicate slack variable, xiIndicate that training sample, n indicate training samples number.
Preferably, above-mentioned steps S4: according to weighting Support Vector data description model to the operation risk of distributing net platform region because
Son is handled, and is realized the Risk-warning of the distributing net platform region, is included the following steps:
S41, using the corresponding weighting Support Vector data description model of each classification, calculate the distributing net platform region
Relative distance of the operation risk factor to the corresponding suprasphere of each classification；
Relative distance of the operation risk factor of distributing net platform region described in S42, comparison to each classification corresponding suprasphere
Size, and classification corresponding to the suprasphere with lowest distance value is exported as a result, realize that the risk of distributing net platform region is pre-
It is alert.
Specifically, using the corresponding weighting Support Vector data description model of each classification, given test sample point x is calculated
To the relative distance d of N number of supraspherej:
Wherein,For the High Dimensional Mapping vector of sample point x, CjAnd RjThe centre of sphere and radius of respectively j-th suprasphere.
Then compare djThe size of value, is minimized, and corresponding j is respective classes ownership, it may be assumed that
Classification corresponding to suprasphere with lowest distance value is exported as a result, obtains the risk shape of distributing net platform region
State, power distribution station state are divided into devoid of risk, risky, failure three classes.
Preferably, in above-mentioned steps S1: after obtaining the operation risk factor of distributing net platform region as training sample set, also
Include:
The operation risk factor of the distributing net platform region is pre-processed, at the operation risk factor of the distributing net platform region
Manage into the quantization index value that can be used for modeling analysis and mode input.
As shown in figure 4, the present embodiment passes through analysis distribution operation risk data related data sources, using transformer equipment as core
The heart will come from Production Managementsystem For Electricpower Network, Electric Power Marketing System, power information acquisition system, production real-time managing and control system and outside
Meteorological, environmental data is integrated, and is traced to the source, is obtained by data, cleaned, pre-processed and quality evaluation process constructs distribution transforming information
Library, according to modeling it needs to be determined that basic data quality standard, the definition quality of data and cleaning rule.To acquisition and operation risk
Relevant data start the cleaning processing, and removal wherein repeats, is imperfect, mistake, the abnormal datas such as mutation, and carries out to data item
Numerical value change processing.
It should be noted that being partially unsatisfactory for subsequent modeling by Data Integration process, the basic data quilt that training needs
It rejects, each data item is processed to be the quantization index value that can be used for modeling factors analysis and mode input.
More preferably, the present embodiment is also based on big data platform curing data pretreatment logic, is often preordained by kafka
When extract each related system increment business datum, with timed task realize incremental data online pretreatment, meet the access of T+1
Data more new demand.
Preferably, in the step S1: before obtaining the operation risk factor of distributing net platform region as training sample set, also
Include:
Using group's method of random sampling is divided, distributing net platform region is chosen, then using distributing net platform region as the correlation of sample extraction not homologous ray
Data obtain the business correlation occurred with distribution platform operation risk, the stronger index factor of regularity.It specifically includes:
(1) it formulates and divides group regular, with distributing net platform region location, the time limit stepping that puts into operation, platform area capacity stepping, the mode of connection
Four conditional combinations are as Partition condition；
(2) random sampling in proportion randomly selects typical platform area according to distributing net platform region quantity in group on year-on-year basis；
(3) as unit of selected power distribution station of sampling, respectively from Production Managementsystem For Electricpower Network, Electric Power Marketing System, use
Related data is extracted in power utilization information collection system, production real-time managing and control system and outside weather system, obtains core data and non-
Core data.
This programme is illustrated by taking certain city history distribution net platform region Risk-warning process as an example below:
As shown in figure 3, power distribution station operating status is divided into normally according to certain city history distribution net platform region operating condition
Three (A class), risky (B class), failure (C class) classifications.
It is respectively 3000,900,900 that A, B, C three classes power distribution station sample are chosen in experiment, randomly chooses training sample
Sheet and test sample, are allocated as follows: training sample is A class 2000, B class 600, C class 600, and test sample is A class 1000
A, B class 300, C class 300.
From state quantity of the equipment, the operation time limit, installation site, power distribution station load factor, power distribution station three-phase imbalance, the external world
State of weather etc. extracts the power distribution station operation risk factor, constructs 10 dimensional feature vector (R1、R2、R3、L1,L2,L3,T1,
T2,T3, W), by principal component analysis, dimension-reduction treatment is carried out to input vector, is described as power distribution station risk status feature, with
This establishes the more disaggregated models of WSVDD.
Then the weight coefficient of the calculation risk factor runs wind using Support Vector data description algorithm building distributing net platform region
The weighting Support Vector data description model of danger.
The operation risk of distribution net platform region is predicted using weighting Support Vector data description model.
By the relevant risks and assumptions of substation equipment, the distribution net platform region load factor factor and distribution net platform region three-phase in platform area
Three characteristic variables of unbalance factor, combination of two are input to weighting Support Vector data description model, obtained result such as Fig. 5
To shown in Fig. 7.
As can be seen that the more disaggregated models of power distribution station risk status WSVDD generally substantially can be right in from Fig. 5 to Fig. 7
Platform area normal sample, risk sample, fault sample are made accurate description and are distinguish.However due to the otherness of different characteristic,
Cause platform area sample different in the identification susceptibility of different characteristic dimension.In conjunction with Fig. 5 and Fig. 7, observation discovery three classes platform area exists
Heavy-overload degree characteristic aspect differs greatly, and especially normal platform area and the area Ling Liangleitai are larger compared to deviation；In conjunction with Fig. 6 and figure
From the point of view of 7, it is seen that there is larger difference in risk platform area, CCS casual clearing station area with normal platform area in three-phase imbalance state, but each other
Between be difficult to distinguish, be easy to cause classification judge by accident；Equally, it can be seen that risk platform area to substation equipment in conjunction with Fig. 5 and Fig. 6
Relevant risks and assumptions feature is more sensitive.
From the point of view of quantitative analysis, the power distribution station more classification results of risk status WSVDD are counted, are classified as follows
Confusion matrix such as table 1:
Table 1
| Classification | A class | B class | C class | It is total |
| A class | 809 | 104 | 87 | 1000 |
| B class | 15 | 262 | 23 | 300 |
| C class | 20 | 21 | 259 | 300 |
From the point of view of the classification results of table 1, WSVDD disaggregated model overall classification accuracy rate is 83.13%, Kappa coefficient
It is 0.7099, method feasibility is stronger.
Meanwhile to verify superiority of the more disaggregated models of this paper WSVDD in the risk identification of power distribution station, by above-mentioned experiment
Sample using common k nearest neighbor, SVM it is more classification method compares experimental analysis, the results are shown in Table 2:
Table 2
| Experimental method | Accuracy rate | Kappa coefficient |
| K nearest neighbor | 78.65% | 0.6225% |
| SVM classifies more | 81.52% | 0.6894% |
| WSVDD | 83.13% | 0.7099% |
From experimental result as can be seen that using the more classifying identification methods of WSVDD, under the same conditions, classification accuracy and
Kappa coefficient is above traditional more classification methods of k nearest neighbor and SVM, i.e. identification power distribution station overall effect is more superior, especially
It is to identify risk platform area and when CCS casual clearing station area, it is more sensitive, be conducive to that distribution operating personnel carries out platform area risk and failure is known
Not, thus for effectively support power distribution station running optimizatin.
As shown in figure 8, the present embodiment additionally provides a kind of distributing net platform region Risk-warning based on Support Vector data description
System, comprising: obtain module 10, dimensionality reduction module 20, model construction module 30 and Risk-warning module 40；
Obtain module 10 and be used to obtain the operation risk factor of distributing net platform region as training sample set, the operation risk because
Attached bag include risks and assumptions relevant to transformer, with risks and assumptions of operation of power networks environmental correclation and relevant with external environment
Risks and assumptions；
Dimensionality reduction module 20 is used to carry out dimensionality reduction to the training sample in the training sample set using Principal Component Analysis
Processing, obtains the training sample of low dimensional；
Model construction module 30 is used for the feature difference of the training sample based on low dimensional, building weighting supporting vector number
According to the more disaggregated models of description；
Risk-warning module 40 is used for the operation according to the weighting more disaggregated models of Support Vector data description, to distributing net platform region
Risks and assumptions carry out feature extraction, the power distribution station of different conditions are identified, to realize the Risk-warning of the distributing net platform region.
It should be noted that the distributing net platform region Warning System based on Support Vector data description disclosed in the present embodiment
In modules for realizing each process in Fig. 1, Fig. 3, the technical effect reached and matching in support vector description
It is identical in the method for prewarning risk of net platform region, it repeats no more at this.
The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all in spirit of the invention and
Within principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.
Claims (10)
1. a kind of distributing net platform region method for prewarning risk based on Support Vector data description, which comprises the steps of:
The operation risk factor of distributing net platform region is obtained as training sample set, which includes related to transformer
Risks and assumptions, to the risks and assumptions of operation of power networks environmental correclation and risks and assumptions relevant with external environment；
Dimension-reduction treatment is carried out to the training sample in the training sample set using Principal Component Analysis, obtains the instruction of low dimensional
Practice sample；
The feature difference of training sample based on low dimensional, the building weighting more disaggregated models of Support Vector data description；
According to the weighting more disaggregated models of Support Vector data description, feature extraction is carried out to the operation risk factor of distributing net platform region,
The power distribution station for identifying different conditions, to realize the Risk-warning of the distributing net platform region.
2. the distributing net platform region method for prewarning risk based on Support Vector data description as described in claim 1, which is characterized in that
The risks and assumptions relevant to transformer include transformer state amount factor R1, transformer station high-voltage side bus time limit factor R2And transformation
Device installation site height above sea level impact factor R3, in which:
R2=eN-M,
R3=e(h-1000)/8150,
In formula, WiIndicate that equipment i relevant with transformer scores recently, aiIndicate equipment i relevant with transformer to transformer
Weighing factor, n indicate equipment sum relevant with transformer；N indication transformer actually puts into operation the time, and M is and transformer
Specified maximum runing time, e is Euler's numbers.
3. the distributing net platform region method for prewarning risk based on Support Vector data description as described in claim 1, which is characterized in that
It is described include power distribution station load factor factor L and distributing net platform region three-phase imbalance with operation of power networks environmental correclation risks and assumptions because
Sub- T, power distribution station load factor factor L include heavy-overload time L in the transformer unit time1, heavy-overload extreme value L in the unit time2
And heavy-overload severity L in the unit time3；
Distributing net platform region three-phase imbalance factor T includes the three-phase imbalance time T of transformer1, transformer three-phase imbalance extreme value
T2With the three-phase imbalance severity T of transformer3。
4. the distributing net platform region method for prewarning risk based on Support Vector data description as described in claim 1, which is characterized in that
The risks and assumptions relevant to external environment include ambient weather state factor W, and ambient weather state factor W includes allocated radio
Temperature, wind-force and the weather condition of environment described in area.
5. the distributing net platform region method for prewarning risk based on Support Vector data description as described in claim 1, which is characterized in that
It is described that dimension-reduction treatment is carried out to the training sample using Principal Component Analysis, obtain the training sample of low dimensional, comprising:
By the risks and assumptions relevant to transformer, with the risks and assumptions of operation of power networks environmental correclation and with external environment phase
The risks and assumptions of pass form multidimensional characteristic vectors；
Dimension-reduction treatment is carried out to the multidimensional characteristic vectors using Principal Component Analysis, obtains the training sample of low dimensional.
6. the distributing net platform region method for prewarning risk based on Support Vector data description as described in claim 1, which is characterized in that
The feature difference of the training sample based on low dimensional, building weighting Support Vector data description model, comprising:
Percentage contribution of the training sample based on the low dimensional when constructing Support Vector data description model, to low dimensional
Corresponding weighted value is arranged in training sample；
In the training sample of the low dimensional and its corresponding weighted value input support vector machines, will construct the weighting support to
Measure data descriptive model.
7. the distributing net platform region method for prewarning risk based on Support Vector data description as described in claim 1, which is characterized in that
It include the sample set of multiple classifications, the feature difference of the training sample based on low dimensional in the training sample set
Property, building weighting Support Vector data description model, comprising:
With the sample set of each classification in the training sample set, the sample set of each classification is constructed respectively accordingly most
Small encirclement suprasphere, as the corresponding weighting Support Vector data description model of each classification, the weighting supporting vector data
Descriptive model are as follows:
Wherein, | | xi-c||2≤R2+ξi, ξi>=0, R indicate that suprasphere radius, C indicate that penalty coefficient, c indicate the suprasphere centre of sphere,
siIndicate weight coefficient, ξiIndicate slack variable, xiIndicate that training sample, n indicate training samples number.
8. the distributing net platform region method for prewarning risk based on Support Vector data description as claimed in claim 7, which is characterized in that
It is described that the operation risk factor of distributing net platform region is handled according to weighting Support Vector data description model, realize the distribution platform
The Risk-warning in area, comprising:
Using the corresponding weighting Support Vector data description model of each classification, the operation risk of the distributing net platform region is calculated
Relative distance of the factor to the corresponding suprasphere of each classification；
Compare the operation risk factor of the distributing net platform region to the corresponding suprasphere of each classification relative distance size, and will
Classification corresponding to suprasphere with lowest distance value exports as a result, realizes the Risk-warning of distributing net platform region.
9. such as the described in any item distributing net platform region method for prewarning risk based on Support Vector data description of claim 1-8,
It is characterized in that, after the operation risk factor for obtaining distributing net platform region is as training sample set, further includes:
The operation risk factor of the distributing net platform region is pre-processed, by the operation risk factor treatment of the distributing net platform region at
It can be used for the quantization index value of modeling analysis and mode input.
10. a kind of distributing net platform region Warning System based on Support Vector data description characterized by comprising obtain mould
Block, dimensionality reduction module, model construction module and Risk-warning module；
Acquisition module is used to obtain the operation risk factor of distributing net platform region as training sample set, which includes
Risks and assumptions relevant to transformer, with the risks and assumptions of operation of power networks environmental correclation and risk relevant with external environment because
Son；
Dimensionality reduction module is used to carry out dimension-reduction treatment to the training sample in the training sample set using Principal Component Analysis, obtains
To the training sample of low dimensional；
Model construction module is used for the feature difference of the training sample based on low dimensional, building weighting Support Vector data description
More disaggregated models；
Risk-warning module is used for according to the weighting more disaggregated models of Support Vector data description, to the operation risk of distributing net platform region because
Son carries out feature extraction, the power distribution station of different conditions is identified, to realize the Risk-warning of the distributing net platform region.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811418180.XA CN109670676A (en) | 2018-11-26 | 2018-11-26 | Distributing net platform region method for prewarning risk and system based on Support Vector data description |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811418180.XA CN109670676A (en) | 2018-11-26 | 2018-11-26 | Distributing net platform region method for prewarning risk and system based on Support Vector data description |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109670676A true CN109670676A (en) | 2019-04-23 |
Family
ID=66143132
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811418180.XA Pending CN109670676A (en) | 2018-11-26 | 2018-11-26 | Distributing net platform region method for prewarning risk and system based on Support Vector data description |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109670676A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110110981A (en) * | 2019-04-26 | 2019-08-09 | 重庆第二师范学院 | A kind of credit rating Default Probability estimates and method for prewarning risk |
| CN110110784A (en) * | 2019-04-30 | 2019-08-09 | 贵州电网有限责任公司 | A kind of transformer fault discrimination method based on transformer correlation operation data |
| CN110458444A (en) * | 2019-08-07 | 2019-11-15 | 国网湖北省电力有限公司计量中心 | A kind of low-voltage platform area operating status multidimensional influence factor integrated evaluating method and system |
| CN110673079A (en) * | 2019-08-23 | 2020-01-10 | 北京中电飞华通信股份有限公司 | Platform district end sensing system and method |
| CN110956331A (en) * | 2019-12-04 | 2020-04-03 | 汇鼎数据科技(上海)有限公司 | Method, system and device for predicting operation state of digital factory |
| CN111479140A (en) * | 2020-03-27 | 2020-07-31 | 深圳市酷开网络科技有限公司 | Data acquisition method, data acquisition device, computer device and storage medium |
| CN112257923A (en) * | 2020-10-21 | 2021-01-22 | 国网冀北电力有限公司承德供电公司 | Heavy overload early warning method and device and electronic equipment |
| CN112258342A (en) * | 2020-10-21 | 2021-01-22 | 国网冀北电力有限公司承德供电公司 | Heavy overload early warning method and device and electronic equipment |
| CN112270622A (en) * | 2020-11-19 | 2021-01-26 | 广东电网有限责任公司电力调度控制中心 | Power distribution network difference characteristic diagnosis and analysis method based on power enterprise big data technology |
| CN113609752A (en) * | 2021-06-28 | 2021-11-05 | 广西电网有限责任公司电力科学研究院 | Distribution transformer loss evaluation system and method for southwest vortex |
| CN113988700A (en) * | 2021-11-15 | 2022-01-28 | 广东电网有限责任公司 | Energy power industry chain risk early warning method and device, terminal equipment and medium |
| CN115276105A (en) * | 2022-09-26 | 2022-11-01 | 国网浙江省电力有限公司宁海县供电公司 | Photovoltaic access capacity planning and multi-energy complementary distributed energy management method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105137238A (en) * | 2015-08-27 | 2015-12-09 | 刘利强 | Fault diagnosis system for gas insulation combination electric appliance |
| CN105894023A (en) * | 2016-03-31 | 2016-08-24 | 华东理工大学 | Support vector data description improved algorithm based on clusters |
| CN106483947A (en) * | 2016-09-21 | 2017-03-08 | 国网江苏省电力公司南通供电公司 | Distribution Running State assessment based on big data and method for early warning |
| CN107220905A (en) * | 2017-05-26 | 2017-09-29 | 国网甘肃省电力公司 | The distribution transformer running status method for prewarning risk that a kind of logic-based is returned |
-
2018
- 2018-11-26 CN CN201811418180.XA patent/CN109670676A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105137238A (en) * | 2015-08-27 | 2015-12-09 | 刘利强 | Fault diagnosis system for gas insulation combination electric appliance |
| CN105894023A (en) * | 2016-03-31 | 2016-08-24 | 华东理工大学 | Support vector data description improved algorithm based on clusters |
| CN106483947A (en) * | 2016-09-21 | 2017-03-08 | 国网江苏省电力公司南通供电公司 | Distribution Running State assessment based on big data and method for early warning |
| CN107220905A (en) * | 2017-05-26 | 2017-09-29 | 国网甘肃省电力公司 | The distribution transformer running status method for prewarning risk that a kind of logic-based is returned |
Non-Patent Citations (3)
| Title |
|---|
| 赵峰等: "配电变压器停运风险的预警分析及研究", 《安徽电力》 * |
| 赵峰等: "配电变压器停运风险的预警分析及研究", 《安徽电力》, no. 03, 30 September 2017 (2017-09-30) * |
| 韩贵金: "基于改进CNN和加权SVDD算法的人体姿态估计", 《计算机工程与应用》 * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110110981A (en) * | 2019-04-26 | 2019-08-09 | 重庆第二师范学院 | A kind of credit rating Default Probability estimates and method for prewarning risk |
| CN110110981B (en) * | 2019-04-26 | 2021-07-20 | 重庆第二师范学院 | Credit rating default probability measure and risk early warning method |
| CN110110784A (en) * | 2019-04-30 | 2019-08-09 | 贵州电网有限责任公司 | A kind of transformer fault discrimination method based on transformer correlation operation data |
| CN110458444A (en) * | 2019-08-07 | 2019-11-15 | 国网湖北省电力有限公司计量中心 | A kind of low-voltage platform area operating status multidimensional influence factor integrated evaluating method and system |
| CN110673079A (en) * | 2019-08-23 | 2020-01-10 | 北京中电飞华通信股份有限公司 | Platform district end sensing system and method |
| CN110673079B (en) * | 2019-08-23 | 2021-12-14 | 北京中电飞华通信股份有限公司 | Platform district end sensing system and method |
| CN110956331A (en) * | 2019-12-04 | 2020-04-03 | 汇鼎数据科技(上海)有限公司 | Method, system and device for predicting operation state of digital factory |
| CN111479140A (en) * | 2020-03-27 | 2020-07-31 | 深圳市酷开网络科技有限公司 | Data acquisition method, data acquisition device, computer device and storage medium |
| CN111479140B (en) * | 2020-03-27 | 2022-07-05 | 深圳市酷开网络科技股份有限公司 | Data acquisition method, data acquisition device, computer device and storage medium |
| CN112257923A (en) * | 2020-10-21 | 2021-01-22 | 国网冀北电力有限公司承德供电公司 | Heavy overload early warning method and device and electronic equipment |
| CN112258342A (en) * | 2020-10-21 | 2021-01-22 | 国网冀北电力有限公司承德供电公司 | Heavy overload early warning method and device and electronic equipment |
| CN112257923B (en) * | 2020-10-21 | 2023-02-03 | 国网冀北电力有限公司承德供电公司 | Heavy overload early warning method and device and electronic equipment |
| CN112258342B (en) * | 2020-10-21 | 2023-02-03 | 国网冀北电力有限公司承德供电公司 | Heavy overload early warning method and device and electronic equipment |
| CN112270622A (en) * | 2020-11-19 | 2021-01-26 | 广东电网有限责任公司电力调度控制中心 | Power distribution network difference characteristic diagnosis and analysis method based on power enterprise big data technology |
| CN113609752A (en) * | 2021-06-28 | 2021-11-05 | 广西电网有限责任公司电力科学研究院 | Distribution transformer loss evaluation system and method for southwest vortex |
| CN113609752B (en) * | 2021-06-28 | 2023-07-25 | 广西电网有限责任公司电力科学研究院 | Southwest vortex-oriented distribution transformer loss evaluation system and southwest vortex-oriented distribution transformer loss evaluation method |
| CN113988700A (en) * | 2021-11-15 | 2022-01-28 | 广东电网有限责任公司 | Energy power industry chain risk early warning method and device, terminal equipment and medium |
| CN115276105A (en) * | 2022-09-26 | 2022-11-01 | 国网浙江省电力有限公司宁海县供电公司 | Photovoltaic access capacity planning and multi-energy complementary distributed energy management method |
| CN115276105B (en) * | 2022-09-26 | 2022-12-27 | 国网浙江省电力有限公司宁海县供电公司 | Photovoltaic access capacity planning and multi-energy complementary distributed energy management method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109670676A (en) | Distributing net platform region method for prewarning risk and system based on Support Vector data description | |
| CN105117602B (en) | A kind of metering device running status method for early warning | |
| CN110097297A (en) | A kind of various dimensions stealing situation Intellisense method, system, equipment and medium | |
| CN111738462B (en) | Fault first-aid repair active service early warning method for electric power metering device | |
| CN108320043A (en) | A kind of distribution network equipment state diagnosis prediction method based on electric power big data | |
| CN106327062A (en) | Method for evaluating state of power distribution network equipment | |
| CN102663537A (en) | Maintenance system of power equipment based on risk assessment | |
| CN206312210U (en) | A kind of status assessing system of Distribution Network Equipment | |
| CN101859409A (en) | Power transmission and transformation equipment state overhauling system based on risk evaluation | |
| CN110322135A (en) | A kind of method for real-time monitoring and system of grid equipment safe operation state | |
| CN102222276A (en) | State maintenance method and device for secondary equipment of power grid | |
| CN108985649A (en) | Timing ambiguity step analysis appraisal procedure based on wind power equipment anomaly analysis data | |
| CN111949939B (en) | Method for evaluating running state of intelligent electric meter based on improved TOPSIS and cluster analysis | |
| CN110309884A (en) | Electricity consumption data anomalous identification system based on ubiquitous electric power Internet of Things net system | |
| CN110084502A (en) | A kind of power quality controlling equipment running status appraisal procedure and device | |
| CN112308425A (en) | Method for constructing distribution transformer health evaluation index system | |
| CN110705859A (en) | PCA-self-organizing neural network-based method for evaluating running state of medium and low voltage distribution network | |
| CN110298567A (en) | The method for determining typical day load curve using integrated energy system energy consumption big data | |
| CN112712256A (en) | Method for evaluating overhaul sequence of low-voltage distribution network equipment | |
| CN111105218A (en) | Power distribution network operation monitoring method based on holographic image technology | |
| CN109409629B (en) | Acquisition terminal manufacturer evaluation method based on multi-attribute decision model | |
| CN114493238A (en) | Power supply service risk prediction method, system, storage medium and computer equipment | |
| CN109784777B (en) | Power grid equipment state evaluation method based on time sequence information fragment cloud similarity measurement | |
| CN108830434A (en) | A kind of needing forecasting method of transformer equipment defect elimination goods and materials | |
| CN116167659B (en) | Carbon market carbon emission data quality evaluation method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |